* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download univERsity oF copEnhAGEn
Survey
Document related concepts
Proteolysis wikipedia , lookup
Lactate dehydrogenase wikipedia , lookup
Nucleic acid analogue wikipedia , lookup
Point mutation wikipedia , lookup
Basal metabolic rate wikipedia , lookup
Peptide synthesis wikipedia , lookup
Butyric acid wikipedia , lookup
Metalloprotein wikipedia , lookup
Specialized pro-resolving mediators wikipedia , lookup
Fatty acid synthesis wikipedia , lookup
Fatty acid metabolism wikipedia , lookup
Glyceroneogenesis wikipedia , lookup
Genetic code wikipedia , lookup
Biosynthesis wikipedia , lookup
Biochemistry wikipedia , lookup
Transcript
university of copenhagen University of Copenhagen Carbohydrate and amino acid metabolism in rat cerebral cortex in moderate and extreme hypercapnia. Quistorff, Bjørn; Folbergrová, J.; Norberg, Karin; K. Siesjö, Bo Published in: Journal of Neurochemistry Publication date: 1975 Document Version Publisher's PDF, also known as Version of record Citation for published version (APA): Quistorff, B., Folbergrová, J., Norberg, K., & K. Siesjö, B. (1975). Carbohydrate and amino acid metabolism in rat cerebral cortex in moderate and extreme hypercapnia. Journal of Neurochemistry, 25(4), 457-462. Download date: 08. May. 2017 Jourrial of Nrurochrrnrsfrp. 1975. Vol. 25, pp. 457-462. Pergamon Prcss. Printed in Great Britain. CARBOHYDRATE AND AMINO ACID METABOLISM IN RAT CEREBRAL CORTEX IN MODERATE AND EXTREME HYPERCAPNIA J. FOLBERCROVA,' K. NORBERG, B. QUISTORFF and B. K. SESO Brain Research Laboratory, E-Blocket, University Hospital, Lund, Sweden, and Department of Biochemistry A, University of Copenhagen, Denmark (Received 28 February 1975. Accepted 13 March 1975) Abstract-The time course of changes in glycolytic and citric acid cycle intermediates and in amino acids was studied in acute and steady state hypercapnia. Experiments on unanaesthetized animals exposed to 10% C 0 2 for lo? 20 and 60s showed that there was a transient decrease in glycogen concentration, progressive increases in glucose-6-phosphate and fructose-6-phosphate and decreases in pyruvate and lactate. During this time the levels of amino acids and Krebs cycle intermediates did not change, except for a small fall in malate at 60s. The results indicate that there was a decrease in glycolytic flux due to an inhibition of the phosphofructokinase reaction. Since the tissue levels of phosphocreatine, ATP, ADP and AMP were unchanged inhibition of phosphofructokinase was probably due to the fall in pH. Anaesthetized animals were exposed to about 5% COz (for 2, 5, 15, 30 and 60 min) or to about 45% C 0 2 (for 5 and 15 min). Except for succinate, which increased, all citric acid cycle metabolites analysed (citrate, a-ketoglutarate, fumarate and malate) decreased with the rise in C0,-tension. The sum of the amino acids analysed (glutamate, glutamine, aspartate, asparagine, alanine and GABA) decreased at extreme hypercapnia. The results suggest that Krebs cycle intermediates and amino acids are partly used as substrates for energy production when there is reduced pyruvate availability due to hypercapnia. It is proposed that amino acid carbon is made available for oxidation via transamination (aspartate aminotransferase reaction) and deamination (glutamate dehydrogenase reaction) and that citric acid cycle intermediates are metabolized following a reversal of reactions usually leading to C 0 2 fixation. PREVIOUS results from this laboratory have shown that hypercapnia reduces the tissue concentrations of pyruvate and lactate as well as of the Krebs cycle intermediates citrate, a-ketoglutarate (a-KG) and malate (FOLBERGROVA et a!., 1972b, 1974a). Although information on changes in succinate and fumarate was not at hand, it was speculated that the increased CO, tensions led to a loss of carbon skeletons from the Krebs cycle. The cause of this hypothetical loss was not apparent in view of the fact that hypercapnic conditions may ravour anaplerotic reactions due to C 0 2 fixation. Since there was a decrease in the sum of the concentrations of glutamate, glutamine and aspartate (FOLBERGROVA et nl., 19748), a shift of carbon atoms from the Krebs cycle to the amino acid pool did not seem likely. However, firm conclusions could not be drawn since the tissue concentrations of alanine and GABA were not measured. The present experiments were undertaken to obtain additional information on changes affecting carboOn leave of absence from Institute of Physiology, Czechoslovak Academy of Sciences, Prague. Abbreviations used: a-KG, m-ketoglutarate; FDH, fructose-1,6-diphosphate; DHAP, dihydroxyacetone phosphate; 3PG, 3-phosphoglycerate; PCr, phosphocreatine; PFK, phosphofructokinase; OAA, oxaloacetate. hydrate and amino acid metabolism in the cerebral cortex in hypercapnia, and to allow defining the mechanisms involved. The following three aspects wcre studied. Firstly, in order to characterize the initial events, unanaesthetized animals were exposed to about 10% CO, and cerebral cortex was obtained for analyses after 10, 20 or 60s of hypercapnia. Secondly, a more complete picture of citric acid cycle changes was obtained by measuring, apart from citrate, a-ketoglutarate and malate, also isocitrate, succinate and fumarate. Since the major part of the citric acid cycle intermediates was measured changes in the size of this pool could be estimated. Thirdly, the previous estimation of amino acid changes, as based on measurements of glutamate, glutamine and aspartate, was extended to include measurements of GABA, asparagine and alanine. Therefore, citric acid cycle intermediates and amino acids were measured in animals exposed to about 5% CO, (for 2, 5, 15, 30 and 60 min) or to about 45% CO, (for 5 and 15 min). The results obtained allow a tentative interpretation of the mechanisms whereby hypercapnia influences substrate utilization in the brain. METHODS The experiments were performed on male Wistar rats that were allowed free access to pellet food and water until 451 458 J. FOLBERGROVA, K. NORBERG, B. QUISTORFF and B. K. SIESJB operation. There were two main groups: unanaesthetized and anaesthetized. The unanaesthetized animals ( 2 0 5 225 g) were provided with indwelling tail artery catheters in a previous operation performed under halothane anaesthesia. When they had recovered completely from this o p eration they were introduced into the plastic cylinder of 1975; NORthe ‘freeze-clamping’ apparatus (QUISTORFF, BERG et al., 1975). The cylinder was initially flushed with room air at a rate of 5 l/min. After allowing a steady period of 15-20 min an arterial blood sample was taken for control of blood gases and pH. The ‘flushing gas-mixture’ was then changed to 20% C 0 2 and 20% O2 in nitrogen, and after about 10s of hypercapnia changed to 10%C 0 2 and 20% 0, in nitrogen. Brain tissue was sampled for analysis by means of ‘freeze-clamping’ following 10, 20 and 60s of hypercapnia. An arterial blood sample was drawn from all animals for determination of Po, and Pco2 at the moment of sampling. In the anaesthetized series anaesthesia was induced with 2-3% halothane to allow tracheotomy and immobilization with tubocurarine chloride. The animals were maintained on 70% N 2 0 and 30% O2 and ventilated to an arterial ,,P of 35-40mm Hg. One femoral artery was cannulated for blood pressure recording and For anaerobic sampling of arterial blood. A skin incision was made over the intact skull bone to accommodate a plastic funnel for freezing of the brain in situ. About 2&30 rnin after the operative procedure, when the animals were at steady state, the composition of the inspired gas was changed to include either 5 or 45% C 0 2 . Animals given 5% C 0 2 were maintained on 25% O2 and 70% N 2 0 . In experiments involving exposure to 45% C 0 2 the C 0 2 concentration of the gas was increased gradually over a 2 rnin period (to counteract a fall in blood pressure). The final composition of the inspired gas was 45% C 0 2 and 30% O2 in N,. Thus, since 45% CO, induces anaesthesia nitrous oxide was withdrawn. Freezing of the tissue in situ (PONT~N et al., 1973) was started 2, 5, 15, 30 and 60 rnin after the inspired CO, tension had been increased in the 5% group and after 5 and 15 rnin in the 45% C 0 2 group. Arterial blood was and haemoglobin condrawn to determine Po,, P, centration. The brains of animals from the unanaesthetized and anaesthetized series were dissected at - 22°C in a refrigerated glove box. A slice of 1-1.5mm thickness was cut out from the surface of the freeze-clamped samples by means of a small buzz saw (QLUSTORFF, 1975). Cortical tissue from this thin slice was used for analysis. From the anaesthetized animals the superior-lateral cortical areas of one hemisphere were cut out. All samples were weighed and extracted with HC1-methanol. The H a methanol extracts were then brought to 0°C and the tissue was extracted twice with perchloric acid for subsequent enzymatic fluorometric analyses of glucose, glycogen, glucose-6-phosphate (G-6-P), fructose-6-phosphate (F-6-P), fructose-1,6-diphosphate(FDP), dihydroxyacetone phosphate (DHAP), 3-phosphoglycerate (3-PG), lactate, pyruvate. citrate, isocitrate, a-ketoglutarate, succinate, fumarate, malate, aspartate, glutamate, glutamine, asparagine, alanine, GABA, ammonia, phosphocreatine (PCr), ATP, ADP and AMP. The methods used were those of LOWRY & PASSONNEAU (1972). The analytical conditions have been described in previous communications from the laboratory (FOLBERGROV~ et al., 1972& 1974a,b). The oxaloacetate concentrations was calculated from the aspartate arnino- transferase reaction, using an equilibrium constant of 6.7 & KORNBERG, 1957). The results were statistically evaluated using two-sample t-test or Aspin-Welch’s test. The following symbols are used:*P < 0.05, f’P < 0.01, $ P < 0001. (-6s RESULTS Unanaesthetized animals One control group and 3 hypercapnic groups (10, 20 and 60s, respectively) with 6 animals in each group were studied. The control values for arterial P,, and Po,were 38.3 0.9 and 98.8 _+ 2.4 nun Hg, respectively (means s.E.M.), In the 3 hypercapnic groups Pco, was 77.6 k 0.7, 72.2 1.8 and 69.4 2.3 mm Hg, respectively, and arterial Po, exceeded 95 mm Hg in each animal. As stated, the unanaesthetized animals were used to evaluate the initial changes in metabolites studied. However, since hypercapnia did not induce significant changes in organic phosphates or amino acids and since the only citric acid cycle intermediate showing a significant change was malate (a decrease from 0.309 OQO6 to 0.252 k 0.006 pmol/g at 60 s, P < 0001) only the glycolytic metabolites will be dealt with. Figure 1 illustrates the pattern of changes in glycolytic intermediates at 10 and 20s (upper panel) and at 60s (lower panel). At 60s, there were no significant changes in glycogen or glucose, highly significant *+ HYPER C A P N IA 50t I tonlrol values Glyc ~nprnollgwet 181 ll%ue + S E M 2 0 0 7 (n = 6 ) 1 I I 1 I I I Gluc G6-P F-6-P F D P DHAP 3-PG Pyr 3 14 0 089 00120 0 096 0 0242 0 035 0074 t o 1 9 30002 t O O O O 4 i0005~0001 9 0 0 2 +OW1 FIG.1. Changes in the concentrations of glycolytic metabolites in the rat cerebral cortex after 10, 20 and 60 s of hypercapnia (glycogen, glucose, glucose-6-phosphate, fructose-6-phosphate, fructose-l,f%diphosphate, dihydroxyacetone phosphate, 3-phosphoglycerate and pyruvate). The values are means k S.E.M. Filled symbols indicate values significantly different from the controls (Pc 005). Carbohydrate and amino acid metabolism in hypercapnia changes in G-6-P and F-6-P, and highly significant decreases in 3-PG and pyruvate. The data obtained at shorter exposure periods demonstrate that there was a transient decrease in glycogen concentration, that the concentrations of G-6-P and F-6-P increased during the first 20s, and that the pyruvate concentration was significantly reduced already at 10s. Two sets of results are relevant to the question of the mechanisms of the increases in G-6-P and F-6-P and of the decreases in pyruvate and lactate (not shown), respectively. First, when the hypercapnia was prolonged beyond the 10s period there were further increases in G-6-P and F-6-P whereas the glycogen concentration returned at 60s to normal values. Thus, it appears unlikely that G-6-P and F-6-P accumulated as a result of breakdown of glycogen (see Discussion and FOLBERGROV~ et al., 1974~).Second, the results obtained are pertinent to the mechanisms of the decrease in pyruvate concentration. Thus, the tissue lactate concentration remained unchanged at 10 and 20s but decreased from a control value of 0.85 0.02 to 0.62 & 0.03 at 60 s. Furthermore, the blood lactate and pyruvate concentrations did not decrease during the course of hypercapnia (not shown). Thus, the decrease in tissue pyruvate concentration could neither be explained in terms of a shift in the lactate dehydrogenase reaction, nor were there concentration gradients favouring transport of lactate or pyruvate from tissue to blood. In summary, the results obtained on the unanaesthetized animals showed that during the first minute of hypercapnia there is a transient decrease in glycogen concentration, progressive increases in G-6-P and F-6-P, and decreases in pyruvate and lactate. These events precede changes occurring in citric acid cycle intermediates and amino acids, and are unaccompanied by alterations in levels of PCr, ATP, ADP and AMP. 459 0 5% O L , +- I Control values Pyr I co2 I I I C m Isocilr mpmollgwelO106 0324 0007 30004 e0015 t O O O l 11SsUe f S E M 5 Mlnules A 15 I I I I d-KG Succ 0120 0696 to007 roo18 I I I I Fum Ma1 " .-. OAA 0059 0339 56.105 t O O O L to012 r02 FIG. 2. The pattern of citric acid cycle intermediates (pyruvate, atrate, isocitrate, r-keloglutarate, succinate, fumarate, malate and oxaloacetate) after 5 and 15 rnin of 5 and 45% hypercapnia as compared to the normocapnic situation.The results are means f S.E M. tically significant for the groups exposed to 45% CO, for 5 and 15 min ( P < 001) but not for the 5% CO, groups. The rise in succinate was not sufficient to balance the reduction in concentration of other cycle intermcdiates. Thus, in the 45% CO, groups the size of the pool was reduced from 1.55 0.04pmol/g to 1.29 & O.O6(P < 0.01) and 1.10 f 002 (P < 0.01) pmol/g at 5 and 15 min, respectively. With 5% CO, the pool size appeared to decrease although the changes were Anaesthetized animals not statistically significant. The values were 1.58 & In all animals the body temperature was close to 0.05, 1.44 5 0.07, 1.35 t 004,1.33 f 0.08, 1.36 L0.05 prnol/g in the control and the 3 7 T , the arterial Po, exceeded lOOmm Hg, and the 0.05 and 1.37 blood pressure was 140mm Hg, or higher. In the 5% 2, 5, 15, 30 and 60 rnin groups, respectively. The tissue concentrations of amino acids and CO, series the arterial CO, tension increased from a control value of 37.2 & 1.1 mm Hg to 63.9 & ammonia are illustrated in Table 1. The changes in 1.8, 69-5 & 1.2, 68.7 0.1, 72% & 0.6 and 76.1 & glutamate, aspartate, glutamine and ammonia conet al., 19726, 0.9 in the 2, 5, 15, 30 and 60 rnin groups, respectively. firm our previous results (FOLBERGROV~ With 45% CO, the arterial CO, tension was 310 5 9 1974~).Thus, with 5% CO, there was a gradual fall in glutamate, a transient increase in aspartate and and 334 f 4 at 5 and 15 min, respectively. Figure 2 illustrates the pattern of changes in citric an increase in glutamine at 60 min. With 45% CO, acid cycle intermediates at 5 and 15 rnin of hypercap- glutamate fell by about 4 5 pmol/g, aspartate innia (5 and 45% CO,). With 5% CO, the results creased by 2.3 pmol/g and there was a significant rise obtained after 2, 30 and 60 rnin were very similar in ammonia. GABA or asparagine did not change to those shown for 5 and 15 min. The results confirm significantly with 5 or 45% CO,. There were no significant changes in alanine conthose of our previous reports in showing decreases in citrate, a-KG, malate and oxaloacetate that were centration. Data obtained with 5% CO, suggested proportional to the rise in CO, tension (see FOLBER-that a small fall occurred in hypercapnia. However, since there was no change with 45% CO, we conclude GROVA et al., 19726, 1974~).In addition, thcy show that there was a reduction in fumarate and a small that alanine concentration remains essentially unalrise in succinate. The increase in succinate was statis- tered in hypercapnia. J. FOLBERGROVA, K. NORBERG, B. QUISTORFF and B. K. SIESJO 460 TABLE1. CONCENTRATIONS OF AMINO ACIDS Experimental group Time (min) 0 2 5 5% CO, 15 30 60 0 45% co, 5 15 Glut Gln 1367 *033 12.88 f009 5.67 ’ 611 +025 5 0.28 12.86 +a43 f026 1248 f054 11.927 f028 5.51 f0.37 6.04 f044 5.78 f030 6.431 k0.19 13.19 f023 l03lt f014 8.65t. fO.12 f064 4.59 f0.40 5.53 f026 12.55. 515 AND AMMONIA IN THE CEREBRAL CORTEX IN RATS EXPOSED AT AT DIFFERENT TIMES coz ASP Aspn 382 * 009 0.087 f0004 0088 f0002 4.62t f013 445t i0.17 462* f025 419 f015 3.74 k013 414 f0.16 6.19f +019 6.41f k0.27 0.085 f0002 0.085 f0003 0084 f0.004 0.087 fow 0078 f 0003 0086 k 0006 0692 fo a x GABA 1.89 f004 1.83 f003 1.84 f0.07 1-90 f0.02 1.89 f0.07 1.95 f0.06 2.13 +008 2.12 f0.14 2.30 f009 5 AND 45% Ala 0.470 f0039 0442 i0.023 0.422 i0005 0,396 f0025 0374 f0.018 0397 f0035 0281 f 0043 0251 f004 0.279 f0.038 0244 f0.029 0240 * 0503 Om9 0.615 0253 f.0017 0330 f 0037 0469t fO.033 f0044 048 I kO.027 f0041 0281 fa058 256 f0 5 26.0 f0.3 25.8 105 25.6 f0.8 250 f0.7 24.5 f 1.1 25.2 f0.5 23.4 f0.7 22.4. f0.7 The values are means & S.E.M. The abbreviations are as follows: Glut = glutamate, Gln = glutamhe, Asp = aspartate, Aspn = asparagine, GABA = y-aminobutyric acid, Ala = alanine. The concentrations are expressed in pmol/g wet tissue. The results of Table 1 indicate that there is a decrease in the amino acid pool (C amino acids) during hypercapnic conditions. With 5% C 0 2 this decrease was not significant although the results obtained at 30 and 60 min are suggestive. With 45% C 0 2 the pool size decreased significantly and at 15 min this decrease amounted to 2 8 pnollg. At this time there was also an increase in ammonia. The changes in pool size and ammonia indicate that oxidative deamination of amino acids may have occurred (see Discussion). The initial changes affect glycolytic metabolites, and alterations in citric acid cycle intermediates and amino acids represent secondary events. (2) The citric acid cycle pool is decreased. Thus, the balance between reactions that replenish carbon skeletons to the citric acid cycle and those leading to loss of carbon atoms is shifted. Furthermore, estimation of the size of the amino acid pool, and measurement of ammonia content, indicate that amino acids are probably used as substrates during hypercapnia. When discussing the probable mechanisms involved we will make two assumptions. First, the fact that the metaDISCUSSION bolism of citric acid cycle intermediates and amino The new metabolites measured in the present acids is compartmented in the brain (BALLZS & anaesthetized series were isocitrate, succinate, fumar- CWR, 1973) does not influence the conclusions. ate, alanine, GABA and asparagine. The control Second, changes occurring during exposure to 5 and values for succinate and fumarate are in good agree- to 45% C 0 2 are qualitatively similar and data ment with those reported by GOLDBERG et al. (1966) obtained with 45% C 0 2 can therefore be used to but the value for isocitrate was less than half of that demonstrate changes that may escape detection with given by these authors. In our series the extracts were 5% CO,. We will discuss in turn glycolytic events, not treated with Florosil for the isocitrate analyses and those affecting citric acid cycle intermediates and and the fluoroscent blank was too high to allow satis- amino acids. factory analyses. However, the results were internally Glycolytic events. Of key importance is the fact that consistent and they are reported since they clearly pyruvate decreases already after 10 s of hypercapnia. demonstrate a fall in isocitrate levels in hypercapnia. Such a decrease could occur as a result of decreased Our values for GABA and alanine are in good agree- delivery or increased utilization. In all probability, the ment with those of other workers (see DUFFY et al., second possibility can be excluded. First, the meta1972). bolic rate for oxygen remains unchanged at P,, The results obtained in the present experiments values of 8S90mm Hg (EKLOF et al., 1973). Second, have greatly facilitated the interpretation of those of there is no indication that pyruvate or phosphoenolthe previous series (FOLBERGROVA et al., 1972b, 1974a) pyruvate is removed by accelerated COz fixation. and the combined results allow a tentative description Thus, instead of increasing, the citric acid cycle pool of the sequence of metabolic events occurring in decreased. Furthermore, since the main C0,-fixing hypercapnia. This description must remain somewhat enzyme in the brain is probably pyruvate carboxylase speculative and some aspects will require validation (PATEL, 1974) the rate of C 0 2 fixation could vary with with other methods. However, the pattern is suffi- the pyruvate concentration, rather than with the C 0 2 ciently coherent to be presented as a framework on tension or the bicarbonate concentration (see CHENG, which subsequent experiments can be founded. Our 1971). Third, analyses of the alanine concentration interpretation is based on the following two facts. (1) give no support to the possibility that pyruvate is Carbohydrate and amino acid metabolism in hypercapnia 461 lost due to a shift in the alanine aminotransferase strate supply. However, whether amino acids are reaction. Fourth, the results make it less likely that metabolized via the glutamate dehydrogenase reaclactate or pyruvate are lost from the tissue by means tion (see WEIL-MALHERBE, 1971) or via the reactions of transport to the blood. of the purine nucleotide cycle (LOWENSTEIN, 1972) Accepting that the decrease in pyruvate con- their carbon skeletons enter the citric acid cycle centration is due to decreased delivery we must before being oxidized. Thus, in hypercapnia and in enquire into the possible mechanisms. The present hypoglycaemia the reduced availability of pyruvate results have shown that there is an accumulation of seems to cause a net flux of carbon skeletons from G-6-P and F-6-P during the first minute of hypercap- amino acids to the citric acid cycle, and from the nia. Since this accumulation does not seem to be latter to the glycolytic chain. secondary to glycogen breakdown (see also FOLBER- Although the main metabolic events in hypercapnia GROVA et al.. 1974a) it may signify inhibition of phos- can be satisfactorily explained there are some dissimiphofructokinase (PFK). Unequivocal proof of PFK larities between hypercapnia and hypoglycaemia that inhibition would require that the glycolytic flux could warrant a discussion, and the mechanisms responsible be shown to be decreased (KFSBS, 1957). Since for some of the changes observed must remain specuCMRO2is unchanged, at least at moderate degrees lative. We will discuss the fate of aspartate, a-KG, of hypercapnia (see above) a decreased glycolytic flux alanine, glutamine and GABA. should mean that non-carbohydrate substrates are In h ypoglycaemia and in hypercapnia aspartate oxidized. Since the amino acid changes indicate that accumulates at the expense of glutamate, indicating this may bc the case (see below) we tentatively con- that there is a shift in the aspartate aminotransferase clude that there is .a decreased glycolytic flux due to reaction. In severe hypoglycaemia there is a massive PFK inhibition. This conclusion is entirely in keeping accumulation of aspartate whereas in hypercapnia a with previous observations on the effect of a de- secondary fall in aspartate to normal or subnormal values occurs (cf. FOLBERCROVA et al., 1974~).In both 1965). creased pH on PFK activity (DANFORTH, Changes affecting citric acid cycle intermediates and conditions the accumulation of aspartate may be amino acids. A decrease in pyruvate concentration due secondary to a relative or absolute increase in OAA. to diminished production implies that a condition of Further metabolism of aspartate would either require ‘substrate depletion’ is at hand. Since CMR,, is un- a reversal of the aspartate aminotransferase reaction changed, at least at moderate degrees of hypercapnia, and the subsequent deamination of glutamate, or deaa decreased glycolytic flux should mean that non- mination by the reactions of the purine nucleotide carbohydrate substrates are oxidized. In fact, sub- cycle. In hypoglycaemia, transamination could be strate depletion during hypercapnia has some striking blocked by the persisting rise in OAA caused by a similarities with hypoglycaemia (LEWISet al., 1974a; depletion of NADH and a fall in malate/OAA ratio, NORBERG & SIESJO,1975). In both conditions there and metabolism via the purine nucleotide cycle could & SIESJ~, is a decrease in the size of the TCA pool, a shift be inhibited by lack of GTP (see NORBERC in the aspartate aminotransferase reaction towards 1975). Apparently, one or both of these pathways aspartate formation, a decrease of the amino acid operate in hypercapnia. In both hypercapnia and hypoglycaemia a rise in pool, and an increase in ammonia. These events suggest that the non-carbohydrate substrate oxidized OAA may thus be one of the mechanisms eliciting during hypercapnia and hypoglycaemia are at least a shift in the aspartate aminotransferase reaction. in part tricarboxylic acid intermediates and amino However, since a-KG does not accumulate other acids. Normally, carbon skeletons are lost from the reactions must accomplish a-KG removal. In hypercitric acid cycle via synthetic reactions, e.g. these lead- capnia, carboxylation of cr-KG to isocitrate could be ing to formation of GABA and glutamine. This loss one of these reactions (see C ~ E N C ,1971). In fact, a is probably mainly replenished through C 0 2 fixation, reversal of the isocitrate dehydrogenase reaction may, and it has been estimated that about 10% of the phos- in removing a-KG, by itself induce transamination. phoenolpyruvate or pyruvate is diverted towards for- There is thus no need to invoke reductive amination mation of malate or OAA (CHENG,1971; BALAZSet of a-KG to explain why its concentration is mainal., 1973). Thus, a net flux of carbon atoms is main- tained low. Instead, since ammonia accumulates when tained from the glycolytic chain to the citric acid cycle there is a decrease in the pool size of amino acids, and from the latter to the amino acid pools. However, the conditions seem to favour oxidative deamination when less pyruvate is available for oxidation there of glutamate. In hypoglycaemia, there are relatively marked deis probably a relative build-up of OAA that may have two effects. Firstly, by inducing a shift in the aspartate creases in alanine, GABA and glutamine. If carboxyaminotransferase reaction it provides amino acid car- lation of a-KG occurs in hypercapnia lack of a-KG bon for oxidation (see DAWSON,1950). Secondly, by could well explain why alanine is not utilized via the allowing a reversal of the C 0 2 fixation reaction it alanine aminotransferase reaction, and why GABA is makes the citric acid cycle intermediates available for not metabolized. In hypoglycaemia, there appears to oxidation, In itself, the citric acid cycle pool is small be a pronounced redox change in the direction of and would not contribute significantly to the sub- oxidation (see LEWISet al., 1974a). Lack of NADH, 462 J. FOLBERGROVA, K. NORBERG, B. QUISTORFF and B. K. SIESJO CHENGS.-C. (1971) Int. Rev. Neurobiol. 14, 125-157. and accumulation of NAD', could therefore enhance DANFORTH W. H. (1965) in Control of Energy Metabolism GABA metabolism by removing succinic semialde(CHANCE B.,ESTABROOK C. W. & WILLIAMSON J. K., eds.) hyde. Thus, COz fixation at the a-KG level and pp. 287-297. Academic Press, New York. absence of a marked redox change could explain why DAWSON R. M. C. (1950) Biochem. J. 47, 386-395. alanine and GABA are not used as substrates in DUFFY T. E., NELSON S. R. & LOWRY0. H. (1972) J. Neurochem. 19, 959-977. hypercapnia. There is no ready explanation for the L., NORBERG K. & S m o delayed increase in glutamine under hypercapnic con- EKLOFB., LASSENN. A., NILLSON B. K. (1973) Acta physiol. scand. 88, 587-589. ditions. However, one factor of importance may be J., MACMILLAN V. & SIESJOB. K. (1972~) that accumulation of ammonia occurs without deple- FOLBERGROVA J . Neurochem. 19, 2497-2505. tion of ATP and the conditions may therefore favour J., MACMILLAN V. & SIESJOB. K. (19726) amidation of glutamate. In hypoglycaemia, ammonia FOLBERGROV~ J. Neurochem. 19, 2507-2517. accumulation is accompanied by energy failure FOLBERGROV~ J., P O ~ U. N & SIESJOB. K. (1974~)J. (LEWISe t al., 1974b) and glutamine formation could Neurochem. 22, 111S1125. be blocked. FOLBERGROV~ J., LJUNGCREN B., NORBERG K. & SIESJOB. In summary, although the detailed mechanisms are K. (1974b) Brain Res. 80, 265-279. largely unknown the present results indicate that GOLDBERG N. D., PASSONNEAU J. V. & LOWRY0. H. (1966) hypercapnia leads to a relative deficiency of pyruvate J. hid. Chem. 241, 39974003. H. A. & KORNBERG H. L. (1957) Energy Transformaand that the changes observed can be at least partly KREBS tions in Living Matter, pp. 213-285. Springer Verlag, Berexplained by the fact that citric acid cycle intermelin. diates and amino acids are used as alternative subLEWIS L. D., LJUNGCREN B., NORBERG K. & SIESJOB. K. strates. study was supported grants from the Swedish Medical Research Council (projects No 14X-263 and 14X-2179), from the Swedish Bank Tercentenary Fund, from U.S. PHS Grant No. R01 so 7838-05 from NIH, and from the Swedish Board for Technical Development. Acknowiedyements-This REFERENCES B A L ~ ZR. S & CREMERJ. E. (1973) Metabolic Compartmentation in the Brain (BALAzs R. & CREMER J. E., eds.). Macmillan, London. BALAZSR., MACHIYAMA Y. & PATELA. J. (1973) in Metabolic Compartmentation in the Brain (BALAZSR. & CR~ME J.RE., eds.) pp. 5S70. Macmillan, London. (1974~)J. Neurochem. 23, 659471. LEWISL. D., LJUNGGREN B., RATCHESONR. A. & Smsio B. K. (19746) J. Neurochern. 23, 673479. bWENSnIN J, M, (1972) Physiol, 52, 382414, L~~~~ 0, H. & pAsSONNEAU J, V. (1972) A Flexible system of Enzymatic Analysis. Academic Press, New York. NORBERG K. & sIEsJii B, K, (1975) Neurochrm. in press. NORHERG K., QUISTORFF B. & SIESJOB. K. (1975) Acta Physiol. scand. Submitted for publication. PATELM. S. (1974) J. Neurochem. 22. 717-724. PONTBNU., RATCHESONR. A., SALFORD L.-G. & SIESJO B. K. (1973) J . Neurochem. 21. 1127-1138. QUISWRFF B. (1975) Analyt. Biochem. In press. WEIL-MALHERBE H. & G O R D ~ N J. (1971) J. Neurochem. 18, 1659-1672. J.